CAGUAS, PUERTO RICO — Miliana Montanez cradled her mother’s head as she lay dying on the floor of her bedroom here, gasping for air and pleading for help.
Ivette Leon’s eyes bulged in terror as she described to her daughter the tiny points of light that appeared before her. She took one last desperate gulp of air just as paramedics arrived. Far too late.
More than eight months after Hurricane Maria devastated Puerto Rico, the island’s slow recovery has been marked by a persistent lack of water, a faltering power grid and a lack of essential services — all imperiling the lives of many residents, especially the infirm and those in remote areas hardest hit in September.
A new Harvard study published Tuesday in the New England Journal of Medicine estimates that 4,645 deaths can be linked to the hurricane and its immediate aftermath, making the storm far deadlier than previously thought. Official estimates have placed the number of dead at 64, a count that has drawn sharp criticism from experts and local residents and spurred the government to order an independent review that has yet to be completed.
The Harvard findings indicate that health-care disruption for the elderly and the loss of basic utility services for the chronically ill had significant impacts, and the study criticized Puerto Rico’s methods for counting the dead — and its lack of transparency in sharing information — as detrimental to planning for future natural disasters. The authors called for patients, communities and doctors to develop contingency plans for such disasters.
Researchers in the mainland United States and Puerto Rico, led by scientists at the Harvard T.H. Chan School of Public Health and Beth Israel Deaconess Medical Center, calculated the number of deaths by surveying almost 3,300 randomly chosen households across the island and comparing the estimated post-hurricane death rate to the mortality rate for the year before. Their surveys indicated that the mortality rate was 14.3 deaths per 1,000 residents from Sept. 20 through Dec. 31, 2017, a 62 percent increase in the mortality rate compared with 2016, or 4,645 “excess deaths.”
The Harvard study's statistical analysis found that deaths related to the hurricane fell within a range of about 800 to more than 8,000, settling on a midpoint overall estimate of 4,645. In response to the survey, the Puerto Rico government released data on June 1 indicating that there had been 1,397 more deaths across the island in the months after the hurricane than during the same period in 2016, though it did not indicate whether those deaths were related to the hurricane.
“Our results indicate that the official death count of 64 is a substantial underestimate of the true burden of mortality after Hurricane Maria,” the authors wrote.
Carlos R. Mercader, executive director of the Puerto Rico Federal Affairs Administration, said in a statement Tuesday that the territorial government welcomes the new Harvard survey and looks forward to analyzing it.
“As the world knows, the magnitude of this tragic disaster caused by Hurricane Maria resulted in many fatalities,” Mercader said. “We have always expected the number to be higher than what was previously reported.”
He said such studies — including a forthcoming George Washington University probe into hurricane fatalities — will help Puerto Rico better prepare for disasters and prevent the loss of life.
Maria, which caused $90 billion in damage, was the third-costliest tropical cyclone in the United States since 1900, the Harvard researchers said.
They also said that timely and accurate estimates of death tolls are critical to understanding the severity of disasters and targeting recovery efforts. And knowing the extent of the impact “has additional importance for families because it provides emotional closure, qualifies them for disaster-related aid and promotes resiliency,” they said.
The researchers noted that the Centers for Disease Control and Prevention says that deaths can be directly attributed to storms such as Maria if they are caused by forces related to the event, whether it is flying debris or loss of medical services.
“The worst part was knowing I could do nothing to help her,” said Leon’s daughter, Montanez, a 29-year-old mother of two. “Knowing she didn’t die peacefully means I will never have closure.”
Puerto Rico’s government faced immediate scrutiny after initially reporting that 16 people had died as a result of the storm, which strafed much of the island Sept. 20. That number more than doubled after President Trump visited in October, when he specifically noted the low death toll. The number kept rising until early December, when authorities said 64 had died.
The official toll included a variety of people from across Puerto Rico, such as those who suffered injuries, were swept away in floodwaters or were unable to reach hospitals while facing severe medical conditions. One was a person from the city of Carolina who was bleeding from the mouth but could not reach a hospital in the days after the storm. After arriving, the patient was diagnosed with pneumonia and died of kidney failure. Another, from Juncos suffered from respiratory ailments and went to the hospital — only to be released because of the coming storm. That person later returned, dead.
The new study indicates there probably were thousands more, like Leon, who died in the weeks and months that followed the storm but were not counted. Their deaths have long raised questions about the manner and integrity of the Puerto Rico government’s protocols for certifying hurricane-related deaths.
Gov. Ricardo Rosselló’s administration did not immediately release mortality data, nor did officials provide much information publicly about the process officials were using to count the dead. But officials and physicians acknowledged privately that there were probably many, many more deaths, and bodies piling up in morgues, across the island.
After pressure from Congress and statistical analyses from news organizations that put the death toll at higher than 1,000, Rosselló enlisted the help of George Washington University experts to review the government’s death certification process. He promised that “regardless of what the death certificate says,” each death would be inspected closely to ensure a correct tally.
“This is about more than numbers,” Rosselló said at a news conference in late February. “These are lives — real people, leaving behind loved ones and families.”
Lynn Goldman, dean of GWU’s Milken Institute School of Public Health, expects an initial report to be released in coming weeks. The school’s findings will include the first government-sponsored attempt by researchers and epidemiologists to quantify Hurricane Maria’s deadliness. Experts are assessing statistical mortality data and planned to dive into medical records and to interview family members of those who have died.
Some cases are obviously storm-related, Goldman said, such as someone dying after a tree branch falls on his head while clearing debris, or someone who suffers a heart attack during the storm and is unable to get help. But death certificates bearing the phrase “natural causes” will require further investigation.
The Center for Investigative Journalism in Puerto Rico has gone to court in an effort to seek the island’s mortality data for the months since November, the last month the government of Puerto Rico shared mortality data publicly. The Puerto Rico Institute of Statistics also announced in recent weeks that it would perform an independent death count and use subpoena powers to retrieve the data.
Spokesman Eric Perlloni Alayon said in a statement that the government is still trying to verify the death toll and does not plan to release any new data.
The Harvard researchers reported that there are several reasons the death toll in Puerto Rico has been drastically underestimated. Every disaster-related death, they said, must be confirmed by the government’s Forensic Sciences Institute, which requires that bodies go to San Juan or that a medical examiner travel to the local municipality.
“As the United States prepares for its next hurricane season, it will be critical to review how disaster-related deaths will be counted, in order to mobilize an appropriate response operation and account for the fate of those affected,” the authors wrote.
Many families here are awaiting clarity on what happened to their loved ones when “natural causes” became the only explanation. That is what was written on Leon’s death certificate. The Puerto Rico Department of Justice’s Yamil Juarbe said in a statement that it is customary for local officials in these cases to review bodies for any signs of trauma and talk to relatives to learn about the deceased’s medical history. That information is collected and sent to the central office of the Forensic Sciences Institute.
Leon’s family said that her name was misspelled on the death certificate and that her death was incorrectly attributed to diabetes; they say she did not have any known chronic diseases. Officials later corrected the documents.
After falling ill while delivering donations to people who lost their homes in a nearby city, Leon sought treatment at Auxilio Mutuo, a private hospital in San Juan. The hospital never lost water service or electricity, said hospital spokeswoman Sofia Luqui, and the 600-bed facility experienced higher-than-usual patient volume after several other hospitals were forced to close.
Leon was diagnosed with diverticulitis and was sent home with prescription drugs, but she did not improve. Montanez said that at 7 a.m. the following morning, her father summoned her to the family home because Leon was short of breath. She died not long after.
Montanez tried for days to have an autopsy performed, but she said no government agency or private medical organization had the capacity to conduct one. Per her wishes, Leon was cremated a few days later in a rushed ceremony because the funeral home was damaged by the storm and was facing an influx of bodies.
Montanez stays awake many nights replaying her mother’s last days. She tries to remember the woman who loved to make wry jokes, who gave each of her neighbors a whistle to call for help in an emergency during the prolonged blackouts, who organized trick-or-treating by lantern light for the children in the barrio after the hurricane.
But mostly Montanez thinks about the storm. The darkness. The lack of services.
“Everything failed. From Day One, everything was failing,” Montanez said. “There are many stories like ours.”
Clarification: An earlier version of this story reported that a Harvard study estimated that there were at least 4,645 deaths attributable to Hurricane Maria and its aftermath. While the study reported that its estimate likely was low, its statistical analysis found that deaths related to the hurricane fell within a range of about 800 to more than 8,000. The story has been updated.
McGinley reported from Washington.